Skip to content

Issues: jameswdelancey/llama3.c

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

8 Open 8 Closed
8 Open 8 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

This could run faster
#17 opened Nov 29, 2024 by ryao
train.py?
#16 opened Nov 23, 2024 by ramezrafla
llama3.1 (and other models) support
#15 opened Aug 4, 2024 by ekk1
1
When non-ascii utf8 chars are passed in the chat input, the reply is always the same.
#9 opened May 6, 2024 by jameswdelancey
Calling the run binary from another program and passing stdin/stdout is hit and miss. Better to have a C ABI exposed via shared library.
#7 opened May 5, 2024 by jameswdelancey
Since llama3 uses other token delimeters than llama2, a switch should be added so that both chat and generate can be tuned to work with both models.
#6 opened May 5, 2024 by jameswdelancey
Tokenization is using some approximation that hasn't been fully tested outside of typical Roman character set
#2 opened May 3, 2024 by jameswdelancey
UNICODE decode of tokens doesn't work quite right in Windows
#1 opened May 3, 2024 by jameswdelancey
ProTip! Follow long discussions with comments:>50.